3D models of humans are commonly used within computer graphics and vision, and so the ability to distinguish between body shapes is an important shape retrieval problem. We extend our recent paper which provided a benchmark for testing non-rigid 3D shape retrieval algorithms on 3D human models. This benchmark provided a far stricter challenge than previous shape benchmarks. We have added 145 new models for use as a separate training set, in order to standardise the training data used and provide a fairer comparison. We have also included experiments with the FAUST dataset of human scans. All participants of the previous benchmark study have taken part in the new tests reported here, many providing updated results using the new data. In addition, further participants have also taken part, and we provide extra analysis of the retrieval results. A total of 25 different shape retrieval methods.
@article{arxiv.2003.08763,
title = {Shape retrieval of non-rigid 3d human models},
author = {David Pickup and Xianfang Sun and Paul L Rosin and Ralph R Martin and Z Cheng and Zhouhui Lian and Masaki Aono and A Ben Hamza and A Bronstein and M Bronstein and S Bu and Umberto Castellani and S Cheng and Valeria Garro and Andrea Giachetti and Afzal Godil and Luca Isaia and J Han and Henry Johan and L Lai and Bo Li and C Li and Haisheng Li and Roee Litman and X Liu and Z Liu and Yijuan Lu and L Sun and G Tam and Atsushi Tatsuma and J Ye},
journal= {arXiv preprint arXiv:2003.08763},
year = {2020}
}